Inverse design of a nano-photonic wavelength demultiplexer with a deep neural network approach

In this paper, we propose a pre-trained-combined neural network (PTCN) as a comprehensive solution to the inverse design of an integrated photonic circuit. By utilizing both the initially pre-trained inverse and forward model with a joint training process, our PTCN model shows remarkable tolerance to the quantity and quality of the training data. As a proof of concept demonstration, the inverse design of a wavelength demultiplexer is used to verify the effectiveness of the PTCN model. The correlation coefficient of the prediction by the presented PTCN model remains greater than 0.974 even when the size of training data is decreased to 17%. The experimental results show a good agreement with predictions, and demonstrate a wavelength demultiplexer with an ultra-compact footprint of 2.6x2.6 mu m(2), a high transmission efficiency with a transmission loss of -2dB, a low reflection of -10dB, and low crosstalk around -7dB simultaneously. (C) 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

Automated summary

In this paper, a pre-trained-combined neural network (PTCN) is proposed as a comprehensive solution for the inverse design of an integrated photonic circuit. The PTCN model shows remarkable tolerance to the quantity and quality of the training data by utilizing both the initially pre-trained inverse and forward model with a joint training process.